Designation: E319 − 85 (Reapproved 2014)

Standard Practice for the

Evaluation of Single-Pan Mechanical Balances1
This standard is issued under the fixed designation E319; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.

INTRODUCTION

The balance performs two basic functions: (1) it compares an unknown load with one or more
weights, and (2) it indicates the difference between the two loads for differences smaller than the
smallest weights normally used on the balance. The test procedure given herein measures the precision
with which the balance can compare the two loads, and the rates at which systematic errors may affect
the observed difference.

1. Scope 2.1.1 accuracy—the degree of agreement of the measure-

1.1 This practice covers testing procedures for evaluating
the performance of single-arm balances required by ASTM
standards.
1.2 This practice is intended for but not limited to sensitivity
ratios of 106 or better and on-scale ranges of 1000xd or more
where d = reability either directly or by estimation.
1.3 This practice can also be applied to other single-pan
balances with mechanical weight changing of different capaci-
ties or sensitivities with appropriate test loads and calibration
weights.
NOTE 1—Mechanical balances of this type have largely been replaced
by automatic electronic balances incorporating a variety of operational
principles. Nevertheless, some single-pan mechanical balances are still
manufactured and many older balances will remain in service for years to
come. One type of automatic electronic balance, the so-called “hybrid,”
bears considerable similarity to single-pan mechanical balances of the null
type. (1)2
1.4 This standard does not purport to address all of the
safety problems, if any, associated with its use. It is the
responsibility of the user of this standard to establish appro-
priate safety and health practices and determine the applica-
bility of regulatory limitations prior to use.
2. Terminology
2.1 Definitions(1):
1
This practice is under the jurisdiction of ASTM Committee E41 on Laboratory
Apparatus and is the direct responsibility of Subcommittee E41.06 on Laboratory
Instruments and Equipment.
Current edition approved Nov. 1, 2014. Published November 2014. Originally
approved in 1968. Last previous edition approved in 2008 as E319 – 85 (2008).
DOI: 10.1520/E0319-85R14.
2
The boldface numbers in parentheses refer to the list of references at the end of
this practice.
ments with the true value of the magnitude of the quantity
measured (2).
2.1.2 correction for a weight—the correction for the error in
adjustment is:
Cr.W 5 A 2 N (1)
where:
Cr.W = correction for the error in adjustment to nominal
value,
A = actual value of the weight, and
N = nominal value.
NOTE 2—In practice it is not possible to adjust weights exactly to their
nominal values.
2.1.3 correction for error in scale indication— the correc-
tion for the scale indication, I, is:
Cr.I 5 A 2 I (2)
NOTE 3—The correction for the scale is taken with reference to the
measured value of a weight used as a test load during calibration of the
on-scale range.
2.1.4 index of precision—the standard deviation, computed
in any acceptable manner, for a collection of measurements
involving a given pair of mass standards (3).
NOTE 4—The standard deviation is computed from the data provided by
the instrument precision test (see Section 7) index of precision.
2.1.5 null-type balance—a balance which requires, as the
final step in its operation, that the observer restore the angle of
the balance beam to its original (or null) position. The least
significant figures of the balance indication are obtained from
this operation.
2.1.6 optical-type balance—in this type the least significant
figures of the balance indication are related to the deflection
angle of the beam from its original (or null) position. A scale
placed on the moving beam is optically projected onto the
(stationary) balance case to provide this indication.

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States

1
 E319 − 85 (2014)
2.1.7 precision—the repeatability of the balance indication
with the same load under essentially the same conditions.
NOTE 5—The more closely the measurements are grouped, the smaller
the index of precision will be. The precision must be measured under
environmental conditions that represent the conditions under which the
balance is normally used.
2.1.8 readability—the value of the smallest decimal subdi-
vision of a scale division in terms of mass units, that can be
read, when the balance is read in the intended manner.
NOTE 6—The readability of a particular instrument is not a measure of
its performance as a weighing device. The relationship between the
numerical value obtained by reading devices and the ability of the operator
to estimate the location of the reference device or index is important. It is
possible to introduce a large number of readable subdivisions of the main
scale divisions that would increase the “readability” as defined but if the
reading device cannot be reset to the same numerical value when the beam
is in an immovable condition, or when the load on the beam is a constant
value, the readability becomes meaningless. Readability substantially less
than 1 standard deviation as determined by repeated measurement with a
given test weight is usually superfluous.
2.1.9 scale division—the smallest graduated interval subdi-
vided either by estimation or with the aid of a vernier.
Subdivisions which appear as divisions on the vernier are not
considered to be scale divisions, but rather parts of scale
divisions.
2.1.10 sensitivity weight—a small weight used to measure
the “on-scale” deflection of the balance indicator.
NOTE 7—With single-pan balances the sensitivity weight should be
equal to the value of the smallest built-in weight represented by the first
step on the dial for the smallest weights.
2.1.11 test load—a load chosen to represent the sample load
in the test procedure.
2.1.12 value of the division—the change in load required to
change the balance indication by one scale division. The
reciprocal of the sensitivity is its most useful function.
3. Summary of Practice
3.1 The accuracy of the direct-reading scale, the smallest
weight of the set of built-in weights, and uniformity of
sensitivity between the upper and lower halves of full-scale
deflections are verified by preliminary tests.
3.2 Estimates of rate of change of the zero with time, rate of
change in the value of the scale division with time, and a
quantitative measure of the variability or random error are
provided by short tests for precision and bias.
3.3 An overall test of the direct-reading capability is pro-
vided by tests of the built-in weights.
4. Significance and Use
4.1 Monitoring Weighing Performance—This practice pro-
vides results in the form of control charts which measure the
weighing capability at the time of the test. A series of tests at
appropriate intervals will monitor balance performance over a
period of time. A marked change from expected performance
may result from a variety of causes including: maladjustment,
damage, dirt, foreign material, and thermal disturbances. If the
2
test results are to indicate future performance, any disturbances
that occur exterior to the balance must be brought under control
(2).
4.2 Acceptance Tests—This practice may also be used as
acceptance tests for new balances. For this purpose, the tests
should be conducted under favorable, but not necessarily ideal,
conditions. Since systematic error in the course of the zero and
the course of the sensitivity may be caused by disturbances
external to the balance, limits on these errors are not ordinarily
prescribed in acceptance requirements.
5. Preparation of Apparatus
5.1 Balance (In all cases, the balance should be used in
accordance with the manufacturer’s instructions):
5.1.1 The results obtained will depend on the environment.
Select an area which is free of excessive vibration and air
currents, where rapid changes in temperature and relative
humidity will not be encountered, and where the floor is rigid
enough to be free of a tilting effect on the balance indication.
Place the balance on a sturdy bench. If the balance has been
moved to a new location, permit it to come to thermal
equilibrium for at least 1 h before performing the test,
preferably several hours.
5.1.2 Inspect and test the balance to make sure that it is in
proper mechanical order. Arrest and release the beam to make
sure that readings are approximately repetitive. Observe the
indication during arrest and release to ensure that there is no
“kick” that would indicate that arrestment points might be out
of adjustment. If necessary, have the balance adjusted by a
competent balance technician.
5.1.3 Make a few trial measurements of the interval from
zero to the full-scale indication.
5.2 Reading the Balance—The balance should be read in
accordance with the instructions supplied by the manufacturer.
Optical types should include the reading of verniers or mi-
crometres. Null types should include the indication of the
device for restoring to null including verniers or micrometres.
6. Preliminary Testing of Single-Arm Balances
6.1 Summary of Method—With single-pan balances the
smallest built-in weight, indicated by the first step on the dial,
is compared with a calibrated weight. The direct-reading scale
is tested for agreement with the smallest built-in weight and the
sensitivity is adjusted, if necessary, so that the indications of
the scale are precise in terms of the calibrated weight. A
“fifty-fifty” test verifies the accuracy of the midpoint at
half-full scale. This test should be performed before proceed-
ing to other tests. After the accuracy of adjustment of the
smallest built-in weight is verified, this weight is used to test
full-scale deflections. Tests are also made for the uniformity of
deflection over the lower and upper halves of the full-scale
deflection. The preliminary tests show either that the balance is
operating properly, or that discrepancies indicate the presence
of sources of error. Uncertainties of perhaps one millionth of
the balance capacity may be caused by dirt or foreign material
in the bearings, or by unskilled handling, while larger discrep-
ancies may be caused by worn or damaged knife-edges or other
sources such as electrostatic effects. Any necessary cleaning or
 E319 − 85 (2014)
servicing should be done at this point. If discrepancies
continue, other possible sources of uncertainty should be
studied. There is no point in proceeding with routine test
procedures until acceptable results can be obtained with the
preliminary tests.
NOTE 8—With null-type balances (including the hybrid) it is possible to
use the flexure of a segment of metal, quartz, etc. as the main pivots
instead of knife edges. A flexure pivot is by its nature free of problems of
dirt. Flexures are also generally more robust than knives. The chief
problem associated with flexures is that they act like springs and thus add
a restoring force which may vary with time or temperature. This drawback
can be minimized by careful design and all but eliminated by the use of
servo-control in electronic balances.
6.2 Materials:
6.2.1 A watch or clock which indicates seconds,
6.2.2 Pencils for recording data,
6.2.3 Columnar data sheets (If balance performance will be
monitored periodically, it may be useful to enter data directly
into a personal computer which has been programmed for this
task.),
6.2.4 A calibrated weight designated S1 which has the
nominal value equal to the smallest interval on the dial-
operated weights, and
6.2.5 Two weights of half of the nominal value of S1
designated (1⁄2)1 and (1⁄2)2. (These weights need not be
calibrated but they should bear distinguishing marks, prefer-
ably one, and two dots.)
6.3 Procedure—Adjust the “no-load” readings to a point
near the center of the vernier so that zero drift or other
deviation will not cause a negative scale reading. Perform the
preliminary tests, loading the pan and changing the dial
settings according to the schedule in Table 1. Before releasing
the beam, record the load on the pan and the dial setting so that
the observation will be confined to the scale reading. Release
the balance and observe the scale reading. Record the indica-
tion and verify the stability of the scale reading, then arrest the
balance promptly.
6.4 Calculations for Preliminary Tests:
6.4.1 Compute D1, the value of the smallest built-in weight
as follows:
D1 5 @ ~ a 2 b1f 2 e ! /2 # 1S1
where: a, b, f, and e are taken from Table 1, and S1 = cali-
brated value of test weight.
6.4.2 Compute S1 in scale divisions to verify the full-scale
value on the direct-reading scale as follows:
S1 5 ~ c 2 b1d 2 e ! /2 (4)
where c, b, d, and e are taken from Table 1. Adjust the
balance sensitivity if necessary so that the full-scale reading
equals D1.
6.4.3 Compute average scale difference, A, for lower 50 %
of direct-reading scale as follows:
A 5 ~ g 2 f1j 2 k ! /2 (5)
6.4.4 Compute average scale difference, B, for upper 50 %
of direct-reading scale as follows:
B 5 ~ h 2 g1i 2 j ! /2 (6)
A and B should agree within 3 standard deviations (see
7.5.3). Any discrepancy smaller than 3 standard deviations may
be ascribed to uncertainty in the preliminary measurements and
does not necessarily indicate a real change in the value of the
scale divisions.
6.4.5 Inspect the no-load readings, a, f, and k for agreement
or zero drift.
6.4.6 See Table 2 and Fig. 1 for examples of calculations
and observation form.
7. Instrument Precision (4)
7.1 Summary of Method:
7.1.1 A set of four readings is repeated four times, or more,
to obtain pairs of readings with identical loads:
7.1.1.1 A reading near zero,
7.1.1.2 A reading near the upper end of the scale,
7.1.1.3 A reading near the upper end of the scale with a test
load plus a small weight, and
7.1.1.4 A reading near zero with the test load but with the
small weight removed.
7.1.2 Readings are taken at a steady pace as rapidly as
practicable, consistent with good practice, and the time is
observed at the start of each set of observations and at the end
of the test.
7.1.3 The balance indications are plotted on a graph to
provide a visual presentation of errors. The zero readings are
(3)
connected to show the course of the zero with time. The
response of the balance to the small weight is plotted. The
course of the sensitivity with time is represented by a plot of

TABLE 1 Schedule for Preliminary Tests of Single-Arm Balances

Observation Time Pan Load Dial Setting Scale Reading
a Record the time zero 0 ...
A
b S1 1 ...
c S1A 0 ...
d Wait 30 s S1A 0 ...
e S1A 1 ...
f Record the time 0 0 ...
g (1⁄2)1B 0 ...
h Add (1⁄2)2B (1⁄2)1 + (1⁄2)2B 0 ...
i Wait 30 s (1⁄2)1 + (1⁄2)2B 0 ...
j Remove (1⁄2)1B (1⁄2)2B 0 ...
k Record the time 0 0 ...
A
S1 = calibrated weight of nominal value equal to the smallest dial-operated weight.
B 1
( ⁄2 ) 1and (1⁄2)2 = weights of nominal value equal to 1⁄2 S1 (not necessarily calibrated but marked for identification).

3
 E319 − 85 (2014)
TABLE 2 Example of Preliminary Test of Direct Reading Balance
Observer: John Doe Date: (May 5)
Observation Time Load Dial Scale Reading
a 1:30 p.m. 0 0 6.4
b S1 1 6.45
c S1 0 1006.5
d 1:34 1⁄2 S1 0 1006.5
e S1 1 6.5
f 0 0 6.4
g 1:39 (500 mg)1 0 506.5
h (500 mg)1 + (500 mg)2 1006.25
i (500 mg)1 + (500 mg)2 0 1006.3
j (500 mg)2 0 506.15
k 0 0 6.4

Calculations:
D1 = [(a − b + f − e) ⁄ 2] + S1 = [(6.4 − 6.45 + 6.4 − 6.5) ⁄ 2 ] + 1000 = 999.925
S1 ( = 1000.006 mg) = [(c − b + d − e) ⁄ 2 ] = [(1006.5 − 6.45 + 1006.5 − 6.5) ⁄ 2] = 1000.025 scale divisions
A = (g − f + j − k) = (506.5 − 6.4 + 506.15 − 6.4) ⁄ 2 = 499.925
B = (h − g + i − j) = (1006.25 − 506.5 + 1006.3 − 506.15) ⁄ 2 = 499.95

FIG. 1 Calculations and Observation Form for Test of Precision

the interval from zero to full-scale angular motion of the beam, 7.1.4 A control chart formed by plotting the range within
or with balances having indication to a null point by a plot of pairs provides a quick estimate of the standard deviation for a
an interval measured on the device for restoring the beam to a single reading. The precision of the balance is estimated by
null indication. computations based on the range within pairs of observations.

4
 E319 − 85 (2014)
7.2 Materials—The following materials are required:
7.2.1 A watch or clock which indicates seconds,
7.2.2 Pencils,
7.2.3 Cross-section paper, 20 lines/in. or 10 lines/cm (If
balance performance will be monitored periodically, it may be
useful to enter data directly into a personal computer which has
been programmed for this task.),
7.2.4 One 50-g test load or other appropriate load (the
correction to nominal value need not be determined),
7.2.5 A sensitivity weight (usually 1 or 0.1 g depending
upon the type of balance) with the nominal value equal to one
interval on the dial for the smallest internal weight. (The
correction for this weight is not needed since the full-scale
indication is compared directly with the smallest weight in the
self-contained set of weights.)
7.3 Preparation:
7.3.1 Draw, in the upper left quarter of the sheet of graph
paper, a horizontal base line for points near zero and a second
horizontal base line for points near full scale (100 or 1000
divisions).
7.3.2 Time is represented in the horizontal direction. Space
observations one square apart and record the time showing
hour, minutes, and seconds at the start of each set of four
readings and at the end of the test. The vertical axis represents
change in balance indication. One square represents the small-
est unit that is intended to be read or estimated. It is, however,
proper to record one half a vernier division as half a square on
the graph.
7.3.3 Refer to Fig. 2 for an example.

FIG. 2 Computation Form for Preliminary Test of Direct-Reading Scale

5
 E319 − 85 (2014)
7.4 Precision Test (Short Test Procedures)—See Table 3 for
summary of test. In the following procedure a 50-g test load is
used as an example. Any other test load can be used with
appropriate dial settings.
7.4.1 The 50-g weight and sensitivity weight equal to one
step on the dial for the smallest weights shall have been in the
balance case for at least 1 h. Load the pan with the sensitivity
weight (1 or 0.1 g for most balances) and leave it on the pan for
the entire series of observations. Close the door. Set the dial for
the smallest weights to “1”. Release the beam and adjust the
zero to a reading in the range from +0.5 to 1 mg so that change
in the zero will not take the indication off scale. Arrest the
beam. After the door has been closed for at least 30 s, proceed
with the following test cycle.
7.4.2 Record the time on the graph. Release the beam at the
time recorded. When the beam comes to rest, observe the
indication and record as a zero reading on the graph. Enclose
the point within a small circle. Look again at the balance
indication to verify the reading. Without observing the time,
but working at a steady pace, continue with the sequence.
7.4.3 Arrest the beam, change the dial from “1” to “0”,
release the beam, read as soon as practicable, record one square
to the right of the first point as a point near full scale (near an
indication of “100” or “1000” divisions as the case may be),
encircle this point, verify the reading, and arrest the beam.
7.4.4 Open the door and place the 50-g test load on the pan
using weight forceps long enough so that your hand is not put
inside the weighing chamber. Shut the door promptly, change
the dial to the appropriate setting, that is “50” or “50.0” leaving
the dial for the smallest weight at “0”. Release the beam, read
and record as a point near an indication of “100” or “1000”
scale divisions, as the case may be. Enclose the point in a small
square to indicate the 50-g load. Arrest and change the dial for
the smallest weight from “0” to “1”. Release, read and record
as a point near zero enclosed in a small square, verify, and
arrest the beam.
7.4.5 Open the door and remove the 50-g test load from the
pan and place it on the floor of the balance case next to the pan
where it will be most convenient to be picked up again. Close
the door. Change the appropriate dial, that is the 10-g incre-
ment dial from “5” to“ 0”. Observe the time estimating a few
seconds ahead to the time the balance will be released.
7.4.6 Repeat the observations as listed from 7.4.4 and 7.4.5
at least three times more to obtain four or more sets of four
readings. Four sets of readings will provide the minimum
number of observations required to treat the data by small
TABLE 3 Summary of Short Test of Precision—Single-Pan Constant-Load Balance
Record starting time on graph.
Step Operation Weight on Pan Dial Setting
1 Record time and release beam, read, record 1 g or 0.1 g
2 Arrest beam, change dial, release beam, read, 1 g or 0.1 g
record, arrest beam
3 Open door, place test load on pan, shut door, set 51 g or 50.1 g
dial, release beam, read, record
4 Arrest, change dial, release, read, record 51 g or 50.1 g
5 Open door, remove 50-g weight, close door, 1 g or 0.1 g
change dial, close door, observe time
Repeat steps 1 to 5 four times for a total of at least 16 points recorded on the graph (see Fig. 2).
Finish with a zero reading (Step 1). Record the time.
6
sample statistics. A more reliable estimate of the standard
deviation may be secured by using more sets of observations.
If desired, additional data can be taken. The total number of
sets of four readings must be an even number so that two
adjacent sets can be paired to provide the difference between
pairs of readings with identical loads. Fig. 2 shows the data and
computations for six sets of four readings arranged so that all
data and computations are on a single sheet of graph paper.
Successive points are to be plotted one square apart, from left
to right.
7.4.7 After completing the desired even number of sets of
four readings, finish with a zero reading. This last reading will
be used to complete the course of the zero with time. Record
the time.
7.4.8 See 7.5 for calculations, and Fig. 2 for an example.
7.5 Calculations—See Fig. 1 for an example.
7.5.1 The Course of the Zero—Connect the points obtained
by Step 1 of the test cycle to show the course of the zero with
time.
7.5.2 The Course of the Sensitivity—Subtract the indications
for the first observation from the second, and the fourth from
the third, and plot to obtain the values of scale difference
corresponding to the first step on the dial for the smallest
weights. Connect these points to show the course of the
sensitivity with time as illustrated in Fig. 1.
7.5.3 Quick Estimate of the Standard Deviation(s)—(4,5) :
Average difference between pairs = (2/π)s = 0.798s
It then follows that s = average difference/0.798
The estimate computed in this manner may not agree with
that computed by more sophisticated procedures if the collec-
tion of data is small. This, however, does not impair its
usefulness since either value may differ from the long range
estimate of the standard deviation computed on sufficiently
large collections of data.
7.5.4 Report on Balance Test—Identifying Data—The bal-
ance should be identified by model and serial number. The
observer, date, and location should be stated. Distinctive
features of the balance should be listed.
7.5.5 Systematic Error—If there was a trend in the data,
state an estimate of the rate of change with time in mass units
per minute, or describe the change in the course of the zero. For
direct reading balances, compare the full scale deflection with
the first dialed interval.
7.5.6 Random Error—State the estimate of the standard
deviation.
Observation Record
1 or 0.1 point near “0” |b. at “0” time near “0” line
0 or 0.0 point near 1000 |b. at time “1” near “1000” or “100”
or 100 line
50 or 50.0 point near 1000 at time “2” near “1000” or “100”
or 100 line
51 or 50.1 point near 0 at time “3” near “0” line
1 or 0.1
 E319 − 85 (2014)
8. Test for Built-In Weights (Single-Pan Constant-Load ate schedules for the various balance models. Observe and
Balance) record the scale readings according to the schedule of test loads
8.1 Summary of Method—In 6.4.2 it was specified that the and dial settings, without readjusting the zero.
sensitivity of the balance is to be adjusted, if necessary, so that 8.3.3 Load the pan with the sensitivity weight equivalent to
the indications of the direct reading scale are accurate. After the full-scale deflection (100 mg in the sample Observation
this adjustment has been performed, a test load, from a set of Schedule, Table 3). Close the door. Set the dial for the smallest
weights whose values are known (5), is compared with two weights to “1”. Record the time, the load on the pan, and dial
combinations of the built-in weights in the smallest decade of setting in the proper columns on the observation sheet. (See
the dial. One dial setting will have the same nominal value as Table 7.) Release the beam at the time recorded. When the
the test load, the other will be smaller by one step on the dial
for the smallest decade. In this manner, suitable test loads are
selected to test all the built-in weights in the smallest decade.
TABLE 6 Example of Report of Test on Build-In Weights
Each decade is tested in a similar manner. (See Table 7)A
8.2 Materials Required—The materials required for this test
are: Balance:.......................... Date:.............................
Serial No:........................ Observer:.......................
8.2.1 Watch or clock which indicates seconds, Capacity: 160 g, digital readout to 0.1 mg, one step on dial 1 g, optical readout
8.2.2 Pencils, 1000 mg.
8.2.3 Columnar data sheets (If balance performance will be A
Various combinations of built-in weights were compared with standards. The
monitored periodically, it may be useful to enter data directly largest indicated deviation of any combination from the nominal value was 0.1 mg.
The corrections for standards were negligible so that they were not taken into
into a personal computer which has been programmed for this account.
task.), and
8.2.4 A set of weights to provide known test loads. (It is
convenient to use weights which have been adjusted close TABLE 7 Test of Built-In Weights
enough to the nominal values so that the corrections are NOTE 1—(Corrections to standard are negligible)
negligible by comparison with the readability of the balance. It
Time Load Dial Scale
may be necessary to account for the errors in adjustment of the
1:42 1g 1 0.6.4
weights). Dial 0 to 1 → 1000.0
div.
NOTE 9—Caution: All direct reading balance indications are in terms of 1g 0 1006.4
apparent mass on some defined or implied basis. The built-in weights of 0 0 6.4
many older single-pan mechanical balances are adjusted on the basis of S2g 2 1006.5
8.4 g/cm3 (that is, the weights, which are usually stainless steel, are j
S1g 3 6.4
adjusted to balance brass standards under normal laboratory conditions). 1000.1
More recently, built-in weights have been adjusted on an 8.0 g/cm3 basis, 0 0 6.4
1:47 S5 4 1006.4
which better approximates the actual density of modern grade S weights j
S5 5 6.4
(6). Using error-free weights adjusted in one basis to check the calibration 1000.0
of weights adjusted in the second basis will require a mass correction of 0 0 6.4
about 0.0007 % (0.7 mg for each 100 g). These corrections are necessary S 20 g 19 1006.4
j
because of the buoyant effect of laboratory air (7,8). S 20 g 20 6.4
1000.0
8.3 Procedure: 0 0 6.4
S 50 g
8.3.1 The set of weights should be placed inside the pan
S 20 g 79 1006.3
compartment of the balance for at least 1 h before starting the S 10 g
j
80 6.3
j
test of the built-in weights. Weights should be in thermal 1000.0
0 0 6.4
equilibrium with the balance. 1g 1 6.4
8.3.2 Adjust the no-load reading to some point near the 1g 0 1006.4
center of the vernier so that variability will not produce a Dial 0 to 1 → 1000.0
div.
negative reading. Table 5 shows a typical test load and
dial-setting schedule. The manufacturer can furnish appropri-

TABLE 4 Example of a Report on a Balance Test

Balance:.......................... Date:............................
Serial No:........................ Observer:.....................
Capacity: 160 g, digital readout to 0.1 mg, one step on the dial 1 g, optical
readout 1000 mg.
Systematic Error: The zero readings increased 0.15 mg in 15 min for an
estimated rate of 10 µg/min. The sensitivity in scale divisions per gram
varied between 1000.2 and 999.95 scale divisions per gram during 18 min.
Random Error: The estimated standard deviation for the difference between a
pair of observations is 0.073 mg.
Comment: On the basis of 12 pairs of duplicate readings taken during an
18-min period, it appears that the systematic error in the balance indication
is relatively unimportant by comparison with the random error.
beam comes to rest, observe the indication and record on the
observation sheet, on the same line as the load and the weight,
but in the column headed “Reading.” Verify the reading.
Without observing time, but working at a steady pace, continue
the sequence.
8.3.4 Arrest the beam, change the dial from “1” to “0”,
release the beam, and record the load and dial settings. Read
the balance indication when the beam comes to rest and record
the indication on the observation sheet, verify the reading, and
arrest the beam.

7
 E319 − 85 (2014)
TABLE 5 Sample Observation Schedules for Test of Built-in Weights
Series of Denominations 80, 40, 20, 10, 5, 2, 1, 1 g Series of Denominations 5, 2,1, 1 g, for
All Decades
Smallest Interval = 1 g Smallesr interval = 0.1 g
Capacity = 160 g Capacity = 100 g
A
Pan Load Dial Setting Scale Reading Pan LoadA Dial Setting Scale Reading
1g 1 ... 100 mg 0.1 ...
1g 0 ... 100 mg 0.0 ...
Zero 0 ... Zero 0.0 ...
S3g 2 ... S 0.3 g 0.2 ...
S3g 3 ... S 0.3 g 0.3 ...
Zero 0 ... Zero 0.0 ...
S5g 4 ... S 0.5 g 0.4 ...
S5g 5 ... S 0.5 g 0.5 ...
Zero 0 ... Zero 0.0 ...
S 20 g 19 ... S3g 2.9 ...
S 20 g 20 ... S3g 3.0 ...
Zero 0 ... Zero 0.0 ...
S 80 g 79 ... S5g 4.9 ...
S 80 g 80 ... S5g 5.0 ...
Zero 0 ... Zero 0.0 ...
1g 1 ... S 30 g 29.9 ...
1g 0 ... S 30 g 30.0 ...
Zero 0.0 ...
S 50 g 49.9 ...
S 50 g 50.0 ...
Zero 0.0 ...
100 mg 0.1 ...
100 mg 0.0 ...
A
Standard weights are designated by S before the denomination.

8.3.5 Open the door, remove the weight from the pan, close 8.4 Calculations—See Table 7 for example in which correc-
the door, release the beam, record the time, load (in this tions to the standard weights are neglected, and Table 8 for an
instance no load), and dial settings. Read as soon as example in which corrections are taken into account.
practicable, record reading, verify reading, and arrest beam. 8.4.1 The value of the weight or combination of weights
8.3.6 Open the door and place the first weight listed in the removed from the balance suspension when the dial is at the
Observation Schedule (in this case, 0.3 g) on the pan. Close the indicated setting may be computed from the following expres-
door, change the dial or dials to the appropriate setting sion:
indicated in the Observation Schedule (in this instance, the dial Value of dialed weight = value of standard (s)
for the smallest weights is set at “2”). Release the beam, record
+ nearest zero reading (a) − load reading (b)
the load and dial settings, read as soon as practicable, record
reading, verify reading, and arrest beam. Example from Table 8:
8.3.7 Change the dials to the next setting indicated in the Dial s a b
Observation Schedule (in this instance the dial for the smallest 0.2 = 300.00 + 0.53 − 100.51 = 200.02 mg
weights from “2” to “3”). Release the beam, record the load 0.3 = 300.00 + 0.54 − 0.53 = 300.01 mg
and dial setting and read as soon as practicable, record reading,
8.4.2 The values of the weights and combinations of
verify reading, and arrest the beam.
weights represented by the various dial settings may be shown
8.3.8 Repeat the observations as listed above from 8.3.5 –
on the observation sheet.
8.3.7 inclusive, for all of the test loads given in the Observation
Schedule. Use appropriate weights and dial settings. After the
last test-load observation, make another zero or no-load obser- 9. Keywords
vation as in 8.3.5 then repeat observations 8.3.3 and 8.3.4. 9.1 balances; scales; weighing instruments

8
 E319 − 85 (2014)
TABLE 8 Example Test of Built-in WeightsA
Balance M-1
Date: Observer:
Observations Computations
Mass of Standard Mass of Dial Weight,
Load Dial Scale Zero Load
Time Weight, mg Nominal + mg Nominal +
on Pan Setting, g Reading, mg ReadingB Reading, mgB
Correction Correction
9:00 100 mg 0.1 0.55 = a 100 mg − 0.004 +0.53 (c) −0.55 (a) = 100 mg − 0.02
100 mg 0.0 100.50 = b
9:02 0 0.0 0.53 = c 300 mg − 0.002 +0.53 (e) −100.51 (d) = 200 mg + 0.02
300 mg 0.2 100.51 = d 300 mg − 0.002 +0.54 (f) −0.53 (e) = 300 mg + 0.01
300 mg 0.3 0.53 = e
9:05 0 0.0 0.54 = f 500 mg + 0.001 +0.54 (f) −100.54 (g) = 400 mg + 0.00
500 mg 0.4 100.54 = g 500 mg + 0.001 +0.50 (i) −0.54 (h) = 500 mg − 0.04
500 mg 0.5 0.54 = h
9:08 0 0.0 0.50 = i 3 g + 0.00 +0.50 (i) −100.51 (j) = 2.9 g − 0.01
3g 2.9 100.51 = j 3 g + 0.00 +0.52 (l) −0.47 (k) = 3.0 g + 0.05
3g 3.0 0.47 = k
9:11 0 0.0 0.52 = l 5 g + 0.02 +0.52 (l) −100.54 (m) = 4.9 g + 0.00
5g 4.9 100.54 = m 5 g + 0.02 +0.48 (o) −0.50 (n) = 5.0 g + 0.00
5g 5.0 0.50 = n
9:14 0 0.0 0.48 = o 30 g − 0.05 +0.48 (o) −100.41 (p) = 29.9 g + 0.02
30 g 29.9 100.41 = p 30 g − 0.05 +0.44 (r) −0.45 (q) = 30.0 g − 0.06
30 g 30.0 0.45 = q
9:17 0 0.0 0.44 = r 50 g + 0.01 +0.44 (r) −100.41 (s) = 49.9 g + 0.04
50 g 49.9 100.41 = s 50 g + 0.01 +0.43 (u) −0.45 (t) = 50.0 g + 0.01
50 g 50.0 0.45 = t
9:20 0 0.0 0.43 = u 100 mg − 0.004 +0.43 (u) −0.42 (v) = 100 mg + 0.01
100 g 0.1 0.42 = v
100 g 0.0 100.36 = w
A
An example of the test of the built-in weights of a semi-micro balance with a readability of 0.01 mg in which the corrections for the weights are taken into account. The
corrections are from NBS 2.6 Test No. 105, Set E. The computations are: mass of the standard weight plus the zero reading minus the load reading equals the mass of
the weights indicated by the dial setting.
B
The letters in parentheses indicate the observation indicated in the column, Scale Reading.

REFERENCES

(1) Schoonover, R. M., “A Look at the Electronic Analytical Balance,”
Analytical Chemistry, Vol 53, 1982, pp. 973A–980A.
(2) Macurdy, L. B., “Response of Highly Precise Balances to Thermal
Gradients,” Journal of Research of the National Bureau of
Standards—C. Engineering and Instrumentation , Vol 68C, No. 3,
July–September 1964.
(3) Eisenhart, Churchill, “Realistic Evaluation of the Precision and
Accuracy of Instrument Calibration Systems,” Journal of Research of
National Bureau of Standards—C. Engineering and Instrumentation,
Vol 67C, No. 2, April–June 1963.
(4) Natrella, M. G., “Experimental Statistics,” National Bureau of Stan-
dards Handbook 91 , October 1966.
(5) Tippett, L. H. C., “On the Extreme Individuals and the Range of
Samples taken from a Normal Population,” Biometrika, Vol 17, 1925,
p. 364.
(6) ASTM E617, Specification for Laboratory Weights and Precision
Mass Standards, Annual Book of ASTM Standards , Vol 14.02.
(7) Jaeger, K. B., and Davis, R. S., A Primer for Mass Metrology,
National Bureau of Standards Special Publication 700-1, August 1984
.
(8) Schoonover, Randall M., and Jones, Frank, “Air Buoyancy Correc-
tions in High-Accuracy Weighing on Analytical Balances,” Analytical
Chemistry, Vol 53, 1981, pp. 900–902.
(9) Macurdy, L. B., “Tests of the Performance of Laboratory Balances,”
Instruments and Control Systems, Vol 38, No. 9, pp. 127–133,
September 1965.

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@astm.org (e-mail); or through the ASTM website
(www.astm.org). Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; http://www.copyright.com/